package com.hughes.android.dictionary.engine;
+import com.hughes.android.dictionary.parser.wiktionary.EnTranslationToTranslationParser;
+import com.hughes.android.dictionary.parser.wiktionary.WholeSectionToHtmlParser;
+import com.hughes.android.dictionary.parser.wiktionary.WiktionaryLangs;
+
+import junit.framework.TestCase;
+
+import java.io.File;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.LinkedHashMap;
import java.util.Map;
import java.util.Set;
-import junit.framework.TestCase;
-
-import com.hughes.android.dictionary.parser.wiktionary.EnTranslationToTranslationParser;
-import com.hughes.android.dictionary.parser.wiktionary.WiktionaryLangs;
-
public class DictionaryBuilderMain extends TestCase {
static final String INPUTS = "data/inputs/";
return isoToStoplist.containsKey(iso) ? isoToStoplist.get(iso) : "empty.txt";
}
+ static String getOtherLang(final String[] pair, final String first) {
+ assert Arrays.asList(pair).contains(first);
+ assert pair.length == 2;
+ return pair[0].equals(first) ? pair[1] : pair[0];
+ }
+
static List<String> getMainArgs(final String[] pair) {
final List<String> result = new ArrayList<String>();
result.add(String.format("--lang1Stoplist=%s", STOPLISTS + getStoplist(lang1)));
result.add(String.format("--lang2Stoplist=%s", STOPLISTS + getStoplist(lang2)));
- int i = 2;
+ int i = 1;
+
+ // For a few langs, put the defs of the other language in DE/IT/FR using WholeSection.
+ for (final String wikitionaryLang : Arrays.asList("EN", "DE", "IT", "FR")) {
+ if (!Arrays.asList(pair).contains(wikitionaryLang)) {
+ continue;
+ }
+ final String foreignIso = getOtherLang(pair, wikitionaryLang);
+ final String wikiSplitFile = String.format("%s/wikiSplit/%s/%s.data", INPUTS, wikitionaryLang.toLowerCase(), foreignIso);
+ if (!new File(wikiSplitFile).canRead()) {
+ System.err.println("Can't read file: " + wikiSplitFile);
+ continue;
+ }
+ result.add(String.format("--input%d=%s", i, wikiSplitFile));
+ result.add(String.format("--input%dName=%sWiktionary.WholeSections.%s", i, wikitionaryLang, foreignIso));
+ result.add(String.format("--input%dFormat=%s", i, WholeSectionToHtmlParser.NAME));
+ result.add(String.format("--input%dTitleIndex=%d", i, Arrays.asList(pair).indexOf(foreignIso) + 1));
+ result.add(String.format("--input%dWiktionaryLang=%s", i, wikitionaryLang));
+ result.add(String.format("--input%dSkipLang=%s", i, foreignIso));
+ ++i;
+ }
// Deal with the pairs where one is English.
if (Arrays.asList(pair).contains("EN")) {
- final String foreignIso = pair[0].equals("EN") ? pair[1] : pair[0];
-
+ final String foreignIso = getOtherLang(pair, "EN");
String foreignRegex = WiktionaryLangs.isoCodeToEnWikiName.get(foreignIso);
if (foreignIso.equals("ZH")) {
foreignRegex = "Chinese|Mandarin|Cantones";
}
- final int enIndex;
- if (foreignIso.equals("DE")) {
- // German-English is a special case since it was the first ever QuickDic!
- result.add(String.format("--lang1=%s", "DE"));
- result.add(String.format("--lang2=%s", "EN"));
- result.add("--dictInfo=@" + INPUTS + "de-en_chemnitz_enwiktionary.info");
-
- enIndex = 2;
- } else {
- result.add(String.format("--lang1=%s", "EN"));
- result.add(String.format("--lang2=%s", foreignIso));
- result.add(String.format("--dictInfo=(EN)Wikitionary-based EN-%s dictionary.%s", foreignIso, getDedication(foreignIso)));
- enIndex = 1;
- }
+ result.add(String.format("--lang1=%s", lang1));
+ result.add(String.format("--lang2=%s", lang2));
+ result.add(String.format("--dictInfo=(EN)Wikitionary-based EN-%s dictionary.%s", foreignIso, getDedication(foreignIso)));
result.add(String.format("--input%d=%s/wikiSplit/en/%s.data", i, INPUTS, foreignIso));
- result.add(String.format("--input%dName=enwiktionary.%s", i, foreignIso)) ;
+ result.add(String.format("--input%dName=ENWiktionary.%s", i, foreignIso)) ;
result.add(String.format("--input%dFormat=enwiktionary", i));
result.add(String.format("--input%dWiktionaryType=EnForeign", i));
result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
- result.add(String.format("--input%dEnIndex=%d", i, enIndex));
+ result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
++i;
result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
result.add(String.format("--input%dWiktionaryType=EnToTranslation", i));
result.add(String.format("--input%dLangPattern=%s", i, foreignRegex));
result.add(String.format("--input%dLangCodePattern=%s", i, foreignIso.toLowerCase()));
- result.add(String.format("--input%dEnIndex=%d", i, enIndex));
+ result.add(String.format("--input%dEnIndex=%d", i, Arrays.asList(pair).indexOf("EN") + 1));
++i;
if (foreignIso.equals("DE")) {
result.add(String.format("--input%dFormat=chemnitz", i));
++i;
}
+
} else {
// Pairs without English.
result.add(String.format("--lang1=%s", lang1));
result.add(String.format("--lang2=%s", lang2));
-
+ result.add(String.format("--dictInfo=Wikitionary-based %s-%s dictionary.", lang1, lang2));
+
result.add(String.format("--input%d=%swikiSplit/en/EN.data", i, INPUTS));
result.add(String.format("--input%dName=BETA!enwiktionary.%s-%s", i, lang1, lang2));
result.add(String.format("--input%dFormat=%s", i, EnTranslationToTranslationParser.NAME));
result.add(String.format("--input%dLangPattern2=%s", i, lang2));
++i;
}
+
return result;
}
// Build the non EN ones.
final String[][] nonEnPairs = new String[][] {
-
+
+ // The 3 I use most:
+ {"IT", "EN" },
+ {"DE", "EN" },
+ {"DE", "IT" },
+
+
{"AR", "DE" },
{"AR", "ES" },
{"AR", "FR" },
{"DE", "IT" },
{"DE", "JA" },
{"DE", "LA" }, // Latin
+ {"DE", "NL" }, // Dutch
{"DE", "PL" }, // Polish
{"DE", "RU" },
{"DE", "SV" }, // Swedish
{"DE", "TR" }, // Turkish
{"DE", "ZH" },
+ {"DE", "TA" }, // Tamil
{"FR", "BG" }, // Bulgarian
{"PL", "HU" }, // Polish
{"PL", "ES" }, // Polish
- //{"TR", "EL" }, // Turkish, Greek
+ {"TR", "EL" }, // Turkish, Greek
+
+ {"FA", "HY" }, // Persian, Armenian, by request.
+ {"FA", "SV" }, // Persian, Swedish, by request.
};
allPairs.addAll(Arrays.asList(nonEnPairs));
}
done.add(pairList);
+ if (!pairList.contains("IT") || !pairList.contains("EN")) {
+ //continue;
+ }
+
DictionaryBuilder.main(getMainArgs(pair).toArray(new String[0]));
}
projects / DictionaryPC.git / blobdiff